Welcome, Guest. Please login or register.
Did you miss your activation email?
Friday 06 December 2024, 08:26:07 am

Login with username, password and session length

The Latest Endian Firewall is now available for download HERE
14261 Posts in 4377 Topics by 6517 Members
Latest Member: Sandro
Search:     Advanced search
+  EFW Support
|-+  Support
| |-+  General Support
| | |-+  EFW starting to work slow after certain time and kernel error
0 Members and 3 Guests are viewing this topic. « previous next »
Pages: [1] Go Down Print
Author Topic: EFW starting to work slow after certain time and kernel error  (Read 12961 times)
vall
Full Member
***
Offline Offline

Posts: 12


« on: Thursday 25 August 2011, 12:23:05 am »

Hello guys,

So...I just installed endian and I'm testing it for a relative small network (500 max clients), i'll start by showing some information about the machine.

Hardware:
Code:
Proc: i5@3.10 GHZ
Ram: 4 GB DDR3
Hdd: 1 TB
NICS: 2 x 1 GB cards

Endian info:
Code:
Appliance	Community
Version 2.4.1
Kernel 2.6.32.25-57.e40.i586
Uptime 10h 19m
Data updated at 17:02:36

Network configuration:
Code:
Green: 10.213.0.1/23, eth0

Red: 10.210.0.213/24, eth1

Services started:
DHCP, firewall, IPS, traffic monitoring

The problem
The problem is the following...after a certain time ( aprox 1 hour) EFW starts to work slow, the internet speed drops from 98 Mpbs to 4 Mpbs and the latencity increases very much.

Here is an example.

Before:
Code:
Pinging 10.213.0.1 with 32 bytes of data:
Reply from 10.213.0.1: bytes=32 time=<1ms TTL=64
Reply from 10.213.0.1: bytes=32 time=<1ms TTL=64
Reply from 10.213.0.1: bytes=32 time=<1ms TTL=64
Reply from 10.213.0.1: bytes=32 time=<1ms TTL=64

After
Code:
Pinging 10.213.0.1 with 32 bytes of data:
Reply from 10.213.0.1: bytes=32 time=66ms TTL=64
Reply from 10.213.0.1: bytes=32 time=65ms TTL=64
Reply from 10.213.0.1: bytes=32 time=64ms TTL=64
Reply from 10.213.0.1: bytes=32 time=64ms TTL=64
Reply from 10.213.0.1: bytes=32 time=62ms TTL=64
Reply from 10.213.0.1: bytes=32 time=61ms TTL=64
Reply from 10.213.0.1: bytes=32 time=61ms TTL=64
Reply from 10.213.0.1: bytes=32 time=59ms TTL=64

the ping starts at 99 ms and goes down to <1 then goes up again to 100 and counting down again...and so on till the end of time Smiley

If I reboot EFW it works fine for about and hour or so.... I have looked into the logs to see what could trigger this and I think the following thing is the problem.

Code:
System	2011-08-24 14:47:20	kernel: [29027.210424] irq 16 nobody cared (try booting with the "irqpoll" option)
System 2011-08-24 14:47:20 kernel: [29027.210428] Pid: 0, comm: swapper Tainted P 2.6.32.25-57.e40.i586 #1
System 2011-08-24 14:47:20 kernel [29027.210430] Call Trace:
System 2011-08-24 14:47:20 kernel [29027.210434] [<c04573fb>] ? __report_bad_irq+0x24/0x69
System 2011-08-24 14:47:20 kernel [29027.210437] [<c0457402>] ? __report_bad_irq+0x2b/0x69
System 2011-08-24 14:47:20 kernel [29027.210439] [<c045752a>] ? note_interrupt+0xea/0x12c
System 2011-08-24 14:47:20 kernel [29027.210441] [<c0456508>] ? handle_IRQ_event+0x1f/0x99
System 2011-08-24 14:47:20 kernel [29027.210443] [<c0457c2c>] ? handle_fasteoi_irq+0x7a/0x96
System 2011-08-24 14:47:20 kernel [29027.210445] [<c0404c86>] ? handle_irq+0x17/0x21
System 2011-08-24 14:47:20 kernel [29027.210447] [<c04044a1>] ? do_IRQ+0x38/0x89
System 2011-08-24 14:47:20 kernel [29027.210449] [<c0403689>] ? common_interrupt+0x29/0x30
System 2011-08-24 14:47:20 kernel [29027.210453] [<c0557194>] ? acpi_idle_enter_bm+0x251/0x287
System 2011-08-24 14:47:20 kernel [29027.210456] [<c05dd36f>] ? cpuidle_idle_call+0x64/0x95
System 2011-08-24 14:47:20 kernel [29027.210458] [<c04020f5>] ? cpu_idle+0x47/0x60
System 2011-08-24 14:47:20 kernel [29027.210461] [<c07dd7ed>] ? start_kernel+0x2c9/0x2cc
System 2011-08-24 14:47:20 kernel [29027.210462] handlers:
System 2011-08-24 14:47:20 kernel [29027.210462] [<f7d60f8e>] (e1000_intr+0x0/0xd8 [e1000])
System 2011-08-24 14:47:20 kernel [29027.210473] Disabling IRQ #16

This error seems to come from a setting or component in the server but I haven't managed to track the respective setting/component responsible yet.

So...anyone willing to trow an idea into this issue?

Thank you!!
Logged
fqureshi
Sr. Member
****
Offline Offline

Posts: 126


« Reply #1 on: Friday 02 September 2011, 08:33:21 am »

Disable outgoing firewall....
Logged
ISilence
Jr. Member
*
Offline Offline

Posts: 3


« Reply #2 on: Friday 07 October 2011, 06:55:24 pm »

I had the exact same problem! But my CPU was Intel Pentium G620 at first. I guess the reason is that the latest hardware we use is not supported by the EFW 2.4.1. I tried to change NICS (1Gb D-Link, 100Mb D-Link, old realtek cards...) but...
The problem was solved using Intel 945 MB, E4600 CPU and DDR2 memory...
Logged
vall
Full Member
***
Offline Offline

Posts: 12


« Reply #3 on: Friday 24 February 2012, 10:58:19 pm »

Almost forgot about this thread Smiley

The problem was from the NIC's, those Intel gigabit cards suffered from overheating and that caused them to go nuts.

Because they are in confined space the heat go to them, we decided to put AC in those rooms and the problem disappeared.

Logged
Pages: [1] Go Up Print 
« previous next »
Jump to:  

Page created in 0.094 seconds with 18 queries.
Powered by SMF 1.1 RC2 | SMF © 2001-2005, Lewis Media Design by 7dana.com